0
$\begingroup$

This topic has been discussed several times and what I present here is one possible solution to deal with the issue of ill-judged voting.

I commented this on another meta post because it was suggested that this community discourages experts/experienced people/specialists because the popular/trivial questions get more attention and votes whereas good, well thought-of and well researched questions get buried. Also the it takes quite some effort and expertise to answer such questions, however, they don't generally attract much votes.

It is understandable that a very focused question may not be understandable by all and it is right to not vote on what you don't understand; it is impossible to make everyone appreciate a good post, however, it is possible to control excessive voting on popular posts.

As I commented on the previously mentioned post, Askfedora (which is another Q/A site modeled on SE), doesn't allow low rep users to vote. If I were to comment in Askfedora-meta about this rule then I would have spoken against it because it is a site that is application oriented and is basically for troubleshooting a problem. Any user who feels someone has provided solution to their problem should be free to appreciate the solution provider. This forum, however, is not about providing solutions (those who generally seek solutions/troubleshooting tips are usually researchers) but is meant to pass on relevant information about different aspects of biology. There are hardly any experiment specific question. In such case it is apt to allow only experienced users to judge if a post is of good quality or not. Such a criterion exists for creating tag synomyms (only those who have enough rep in a tag can vote for tag synonimization).

I propose that a minimum of 100 points be required to upvote/downvote a post. What is your opinion? (there is another problem here: +100 the bonus points because of SE network. But that can be dealt later).

Is this idea even good?

PS: I dont know about askfedora rules at present. This was something like a year and a half back.

$\endgroup$
  • 4
    $\begingroup$ There is already a minimum reputation for voting, 15 for upvotes and 125 for downvotes. $\endgroup$ – Mad Scientist Oct 19 '14 at 12:40
  • $\begingroup$ @MadScientist just checked my Privileges tab, it's 15 for upvotes. Just sayin' :) $\endgroup$ – MattDMo Oct 22 '14 at 4:39
8
$\begingroup$

Votes are a pretty flawed mechanic, the score of a post is much more determined by how many users see it than the quality of the post. Accessibility of the topic is the second major factor, users tend not to vote if they don't understand what the post is about.

I still think that increasing the reputation required to vote would not be a good solution. The major drawback would be that the total volume of votes would likely be drastically reduced. A certain amount of voting is necessary to ensure that enough users have moderation privileges. It is also rather frustrating if the average voting on answers trends towards zero. I did participate in a tag on Stack Overflow a long while back that had a really low amount of voting. It is not encouraging if the best you can usually hope for is one or two votes and the accepted answer mark if you're really lucky, while many good answers stay at a score of zero.

The other majore issue I have with the idea is that I don't think it would actually solve the problem. It would certainly do something, and probably increase the quality of the voting to some extent, but exposure and accessibility would still play major roles in determining the score of posts.

I personally don't think the problems inherent in the current way voting works can be fixed at all, or at least not without causing different problems. But I also don't think it matters that much that voting isn't perfect, it is still good enough in many cases and in my opinion still better than no voting at all.

The main purpose of voting on answers is to make sure that the best answer is at the top. In my observation this succeeds most of the time, not always for sure, but often enough to be useful. The main purpose of voting on questions is to highlight interesting ones and to bury crappy ones. Downvoting crappy ones works reasonably well, highlighting interesting ones one to some extent. A highly voted question is usually an accessible and popular one, not necessarily a good one.

There is one aspect where I personally think it would be nice if something were changed, and that is the influence of the hot questions list displayed network-wide on voting. I've proposed to disallow users with only the association bonus from voting on other sites on MSE, but that was declined by SE. That idea was maybe a bit too extreme and would be rather annoying in practice, but maybe we'll find a better one at some point.

$\endgroup$
3
$\begingroup$

Are we really being motivated by the number under the name? I certainly have noticed the problem that more technical/difficult questions, and occasionally more technical answers, don't receive similar amounts of votes as popular questions. But is that really a problem? I'm a professional researcher and I'm not sticking around because of a number under my name.

Regarding Lab Questions:

I'm just going to ask a question. Often I will end up answering the question myself, but that doesn't bug me. I went looking for the answer and didn't readily find it, now hopefully someone else will.

I think still in the beta stage, we might should do this more often. It's been said many times that trained researchers are trained to go answer their own questions. It's what we do. It's why PubMed has the traffic it does. So if you're an active enough member of the site to be reading the meta, I think you should make an effort to post the questions you find yourself going to PubMed about. And if you find an answer, answer your own question.

Now I can imagine that this might lead to questions of a lower quality, because by the time you properly research the question you probably already have the answer. I again say that this could be a good time to post it to the site anyway. Again, I don't think the rep should be the motivation.

This is the best way I can think of to combat the "negative feedback" loop this site might be subjected to.

Regarding voting on answers:

Maybe I'm more selfish or unusual, but I'm only going to answer a question if it's something I want to write about. I feel as though I'm often motivated to write a good answer for a question so I can refer others to it when I'm asked in real life.

Now I can imagine that we would need a much larger user base to cover all the questions if everyone felt this way. But even if it's on a topic I'm not as interested in writing on, I write an a good answer to provide a good answer not to get voted on.

$\endgroup$
  • $\begingroup$ It is not about the reputation. It is about unnecessary glorification of trivial stuff. Votes should reflect the quality and importance of a post. $\endgroup$ – WYSIWYG Oct 27 '14 at 6:33
  • $\begingroup$ @WYSIWYG Perhaps I misunderstood then. I understand I'm not really utilizing the internal reward system inherent in SE sites. Then would the solution be a concerted effort to down vote popular but not particularly technical questions? $\endgroup$ – Atl LED Oct 27 '14 at 15:49

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .